Adaptive Control Algorithms And Verification Of Video Transmission In Heterogeneous Wireless Networks

With the rapid development of the next generation mobile network and mobile internet,the mobile data traffic has shown explosive growth in recent years,and the throughput of video application in wireless network has increased exponentially in the whole network.In heterogeneous wireless network environment,there are many factors that affect data transmission,such as wireless channel fading,delay jitter,while streaming media transmission has a higher QoE requirement than traditional business,therefore,how to according to the state of network to optimize video transmission,improve the fluency of video playback and level of video quality,so as to ensure that the viewer can obtain higher quality of experience,which is the key of the research on video transmission in wireless network,and a challenging job to provide users with good viewing experience.Based on the current research status of video transmission,aiming at the existing problems of video transmission in heterogeneous wireless networks and how to improve the user viewing experience,this thesis is mainly carried out from the following three aspects:(1)For the problems of the video live broadcast clients cannot adjust video bit rate automatically in the process of video transmission,the wireless video transmission rate adaptive control algorithm based on deep reinforcement learning is proposed in this article,which will consider the end-to-end delay and client buffer occupancy length jointly,establish a feedback control model with the end-toend delay and play cache as the main parameters,and use DQN algorithm to train the wireless video bit rate adaptive transmission control model with the network data and video data sets to get the most suitable video bitrate for video transmission process.The simulation results show that,compared with the bitrate adjustment algorithm based on the client buffer,the proposed algorithm improves the user experience significantly through effective control of delay and cache.(2)For the problem of insufficient bandwidth of single link or poor network environment,which affects the video transmission and playback,this thesis proposes an HD video business-oriented adaptive parallel transmission control method in heterogeneous wireless network,which establishes a control model of parallel link video transmission according to the problems of minimizing the number of links and orderly transmission.The simulation results show that,compared with the single link transmission scenario,the proposed multi-link video transmission algorithm can further improve the stability of video bit rate,end-to-end delay,and playback buffer,thus improving the user experience.(3)The verification platform of QoE-oriented video transmission adaptive bit rate control system is implemented.According to the network parameters and video parameters,the system platform adjusts the level of video bit rate and the playback speed of the client,so as to improve the stability of video transmission and playback,thus improving the user experience.The simulation experiments show that,the system can simulate the scene of live video transmission and improve user experience by using relevant bitrate adjustment algorithm.